I'll say, Experience Builder looks great! Although I hope Esri will continue supporting and even updating WebApp Builder 2D for at least another year or two in active development. It's going to be a large effort migrating dozens of production viewers to a very new architecture. I have some significantly customized WAB viewers that I have for various internal applications. WAB 2D does what it does quite well. I do not need 3D yet. I have done a couple 3D analyses in Pro for our airport and some line of sight commo stuff but no 3D viewers.
React / API 4.x do not solve too many problems I have, at least at the moment. MVC is cool. I can get in to that, I do .NET Core MD Bootstrap for sites. The Popup UI in 4.x will be better though this looks to be a design decision, as most of the things are possible in Dojo and 3.x - like moving popups, tabs, etc. Resizing could also be achieved. I'm wondering about this timeline because I am about to begin planning a major upgrade to 10.7 or 10.8 and Portal; and this is only to get Dashboards and featureLayer service tiling. The latter being a borderline need-to have because of slow performance for heavy layers from many users at once, and Dashboards being a very useful new kind of capability and that I want to host on premise. Therefore I need to focus on back-end next year. WAB took care of the front-end very well for us.
Experience Builder looks like a great compliment to WAB! I would be happy to have that in addition to WAB 2D 3.x continued with active development for another year or two... Jianxia Song do you have a cutoff date for when we'll stop seeing new updates to WAB 2D and/or JS API 3.x?
This is a bigger transition than Arc to Pro, for us in fact. With ArcMap to Pro, it's professionals who understand tools and interfaces change over time and we're ready (soon...maybe next year) to start the transition. Not too worried about that. By then hopefully most of the things in ArcMap will be in Pro. But with web viewers, they have deep integration with other systems and departments and also end users are far less tolerant of upgrading or changing interfaces. There has to be serious business reasons and ROI to 'sell' a change of interface to a user-facing application. When I went from my old Flex viewers to native JS API viewers it was an easy sell, with all of Flash's issues. Native JS API sites to WAB with its vast array of powerful tools was also an easy sell, our sites now can do so much more. The thing is, I've now heavily customized WAB to fix or improve many of its limitations and it has just reached a good state. Meantime, it's been great because the interface for WAB web viewers is now standardized, and the public is familiar with WebApp Builder's interface. I took time to write a detailed manual and YouTube videos for WAB. (example... SAGIS Property Map Viewer )
